Why does the banking industry remain prone to large and costly disruptions despite being so heavily regulated? Is there a need for more regulation, less regulation, or simply different regulation? Our recent Staff Report combines insights from academic research in economics, finance, and accounting to provide a deeper understanding of the challenges involved in designing and implementing bank regulation, as well as opportunities for future exploration. This post focuses on the regulation of bank capital, but the ideas are applicable more broadly.
The Case for Bank Regulation
Consider a simple bank balance sheet, with two types of assets and two types of funding. On the asset side are cash and loans. Cash is fully liquid with zero return; loans are illiquid but generate positive expected returns if held to maturity. On the funding side are deposits and capital. Deposits are short-term debt contracts that promise fixed payments whenever withdrawn; capital is a loss-absorbing equity stake that profits if and only if the realized loan return is higher than expected.
Banks create social value in two main ways, as highlighted in the work that led to the 2022 Nobel Prize in Economic Sciences. First, they make loans to informationally opaque but productive firms that would otherwise struggle to produce. Second, they use the returns from those illiquid loans to create liquid deposit contracts for risk-averse investors. There is a social value to the liquidity service, so it would be suboptimal for the information service to be fully equity-funded. At the same time, loan returns and depositor withdrawals both have some randomness to them, so it would be too risky for the information service to be fully deposit-funded.
Naturally, there are trade-offs when determining how much capital a bank should have. Too little capital means insufficient loss-absorption and an increased risk of insolvency, while too much capital means the bank offers fewer deposits and hence provides a lower liquidity service. A case for regulation emerges when the bank’s evaluation of this trade-off differs from a social planner’s.
A bank chooses capital to maximize its expected profits; the planner chooses capital to maximize social welfare. A large theoretical literature establishes that banks undervalue the loss-absorbing properties of capital relative to the planner and would thus fund themselves with too many deposits in the absence of capital requirements. The root of this undervaluation is that an individual bank does not internalize the negative effects of its failure on other banks. These negative effects are particularly severe in banking because the business model of using loans to back shorter-term claims is susceptible to runs, introducing a role for beliefs that does not exist in other industries. The failure of one bank can trigger panic and lead to failures of other banks. Additional rationales for capital requirements include the destabilizing effects of fire sales when banks try to stave off failure and the possibility of moral hazard when deposit insurance is priced under imperfect information.
At some level, banking is like any other industry where firms that impart externalities—as captured by a difference in the private and social values of firm activity—are regulated by a public agency. Where industries differ is in the size of the externalities that their firms impart, leading to some industries, including banking, being more heavily regulated than others.
How Accounting Discretion Complicates Bank Regulation
A growing empirical literature reviewed in our Staff Report suggests that circumvention of regulatory constraints is particularly pervasive in banking. The simplest theoretical explanation is precisely that regulatory constraints on banks are more binding because the externalities any one bank imparts—and hence the corrective regulations imposed—are larger than for a nonbank firm. Accordingly, banks have a higher marginal benefit of circumventing regulation. It is important to note that regulatory circumvention does not mean violation of regulation; most of the examples in the empirical literature involve banks taking actions that are fully consistent with the letter of the regulation being circumvented. The problem is that one can follow the letter of a regulation without following the spirit, and it is this gap that opens the door to the possibility of actions that loosen the burden of bank regulation. The marginal cost of undertaking such actions is then critical for determining whether regulation will have its intended effect.
Accounting standards can make it easier or harder for a firm to structure its activities one way while also reporting these activities in a way that complies with regulation. Allowing for more discretion in regulatory reporting makes it easier; allowing for less makes it harder. More discretion therefore decreases the cost to a bank of undertaking a given amount of regulatory circumvention. More discretion also decreases the marginal cost of circumventing regulation. In this way, the success of bank regulation is strongly influenced by the discretion that accounting standards afford.
Empirical studies provide many examples where accounting discretion was used to lessen the burden of capital regulation without changing the true nature of bank activity. For example, some thrifts used discretion in loan loss recognition to smooth earnings around the savings and loan crisis of the 1980s, and a number of banks used discretion in the classification of securities to avoid realizing mark-to-market losses in the run-up to the regional banking distress of March 2023. Leading into the 2008 financial crisis, several banks also moved some activities to special-purpose vehicles and used discretion in the reporting of contingent liabilities to guarantee the vehicles just enough that they could be funded on good terms but not so much that the banks incurred substantial capital charges for providing the guarantees.
A potential solution is to limit discretion so that the marginal cost of circumventing regulation exceeds the marginal benefit. However, a long accounting literature argues that there are independent and socially valuable rationales for allowing discretion in financial reporting. For example, discretion enables firms to use their private information to give stakeholders timely signals about fundamentals. Discretion also introduces more dimensions on which managerial performance can be judged, facilitating the design of incentive-compatible contracts and mitigating internal agency problems.
Implications and Open Questions
Academic research on bank regulation and accounting discretion has unfolded largely in parallel, with little attention paid to how these two policy choices interact. What would a more unified approach suggest? Our Staff Report provides a conceptual framework that sheds light on this question.
A meaningful interaction between bank regulation and accounting standards emerges under two conditions. First, more discretion lowers the marginal cost of circumventing regulation. Second, the level of discretion that achieves the benefits detailed in the accounting literature exceeds the level that would eliminate the incentive to circumvent regulation; that is, a tension exists between the objectives of a social planner who chooses both accounting standards and bank regulation. Understanding that discretion interferes with corrective regulation, the planner will choose less discretion than would otherwise be optimal, and understanding that corrective regulation triggers a social cost to discretion (regulatory circumvention), the planner also chooses to implement a lower capital ratio than would otherwise be optimal. The planner will not want to sacrifice all the benefits of accounting discretion if the social cost of a bit of regulatory circumvention is small.
An important direction for future research is modeling discretion and regulation as multidimensional objects. Discretion may be allowed on some parts of the balance sheet but not others, and regulation extends beyond capital requirements since multiple quantities can be regulated. The planner may find combinations of discretion and regulation that are not in tension. For other combinations, the planner will likely have to choose less regulation and less discretion on at least some dimensions. An open question for further research is which dimensions.
Kinda Hachem is a financial research advisor in the Federal Reserve Bank of New York’s Research and Statistics Group.
How to cite this post:
Kinda Hachem, “Designing Bank Regulation with Accounting Discretion,” Federal Reserve Bank of New York Liberty Street Economics, December 15, 2025, https://doi.org/10.59576/lse.20251215
BibTeX: View |
Disclaimer
The views expressed in this post are those of the author(s) and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the author(s).
